SuperAPI AI Directory
Partner with us中文AI newsFavorites

AI agents

3,047 results · Page 72 of 85

OpenShell

AI agents

OpenShell is a secure, private runtime environment from NVIDIA for autonomous AI agents. Its core goal is to let agents read files, install software packages, call APIs, and use credentials, while not giving them unlimited access to your data, keys, or network. Users declare, through policies, what each agent can touch, and OpenShell enforces it. It governs agent behavior in two ways: first, through kernel-level instrumentation, enforcing policy at runtime for every file access, system call, and network connection; second, through formal verification, checking the new access permissions a policy change would allow before it is applied. Each agent runs in an isolated sandbox, with kernel controls restricting accessible files and system calls; all network connections pass through a policy check before leaving the sandbox, agents do not see real credentials, and OpenShell only adds credentials to requests sent to approved endpoints. Formal verification flags high-risk new access before a policy change is approved, pending human review. Running it requires Linux, macOS on Apple Silicon, or Windows with WSL 2 (experimental), as well as Docker, Podman, or host virtualization. The installer sets up the CLI and a local gateway, and the default sandbox image is a minimal Ubuntu with no agent installed.

Read more

Orca-AI-Incident-Archive

AI agents

The Orca AI Incident Archive is an open database dedicated to collecting records of real-world AI agent-related incidents. The data covers January 2025 through October 9, 2026, with a total of 395 AI agent safety incidents arranged by month, plus one precursor record dating back to December 1, 2024. Each record is a Markdown file with a YAML header, containing an attack chain diagram and at least one clickable primary source. Of the 395 records, only 141 have a confirmed victim. The archive's core purpose is to distinguish "agent incidents that actually caused harm" from research demonstrations in which researchers prove that harm is possible. Each record first answers three questions: whether there is a confirmed victim (real_harm), whether AI involvement is confirmed by a primary source (ai_involvement), and whether the entry is an incident, vulnerability disclosure, research demonstration, threat report, or policy action (kind). The body text of the records is in English, titles and summaries are provided in seven languages (English, Chinese, Japanese, Korean, German, French, Spanish), and the full Chinese text of all records is located in the incidents/i18n/zh/ directory.

Read more

ORCH

AI agents

ORCH is an open-source (MIT-licensed) command-line orchestration tool for managing a group of AI agents working in parallel on a terminal or server. It unifies and coordinates various CLI tools such as Claude Code, Codex, Pi, and Cursor, and through mechanisms such as a state machine, isolated git worktrees, automatic retries, inter-department messaging, and mandatory code review, it enables multiple agents to divide labor and collaborate around the same project goal. With just one npm install, no database, cloud service, Docker, or GPU is required, and all state is kept in the local .orchestry/ directory. It is suitable both for independent developers using a small number of agents to advance a project in parallel, and for deploying a complete "department" through preset team templates; it can orchestrate any CLI workflow through shell adapters and supports long-term operation on a server as a headless daemon.

Read more